Sort by
High-resolution acoustic ejection mass spectrometry for high-throughput library screening

An approach is described for high-throughput quality assessment of drug candidate libraries using high-resolution acoustic ejection mass spectrometry (AEMS). Sample introduction from 1536-well plates is demonstrated for this application using 2.5 nL acoustically dispensed sample droplets into an Open Port Interface (OPI) with pneumatically assisted electrospray ionization at a rate of one second per sample. Both positive and negative ionization are shown to be essential to extend the compound coverage of this protease inhibitor-focused library. Specialized software for efficiently interpreting this data in 1536-well format is presented. A new high-throughput method for quantifying the concentration of the components (HTQuant) is proposed that neither requires adding an internal standard to each well nor further encumbers the high-throughput workflow. This approach for quantitation requires highly reproducible peak areas, which is shown to be consistent within 4.4 % CV for a 1536-well plate analysis. An approach for troubleshooting the workflow based on the background ion current signal is also presented. The AEMS data is compared to the industry standard LC/PDA/ELSD/MS approach and shows similar coverage but at 180-fold greater throughput. Despite the same ionization process, both methods confirmed the presence of a small percentage of compounds in wells that the other did not. The data for this relatively small, focused library is compared to a larger, more chemically diverse library to indicate that this approach can be more generally applied beyond this single case study. This capability is particularly timely considering the growing implementation of artificial intelligence strategies that require the input of large amounts of high-quality data to formulate predictions relevant to the drug discovery process. The molecular structures of the 872-compound library analyzed here are included to begin the process of correlating molecular structures with ionization efficiency and other parameters as an initial step in this direction.

Open Access Just Published
Relevant
Artificial intelligence-driven predictive framework for early detection of still birth

Predictive modeling is becoming increasingly popular in the context of early disease detection. The use of machine learning approaches for predictive modeling can help early detection of diseases thereby enabling medical experts to appropriate medical treatments. Stillbirth prediction is a similar domain where artificial intelligence-based predictive modeling can alleviate this significant global health challenge. Despite advancements in prenatal care, the prevention of stillbirths remains a complex issue requiring further research and interventions. The cardiotocography (CTG) dataset is used in this research work from the UCI machine learning (ML) repository to investigate the efficiency of the proposed approach regarding stillbirth prediction. This research work adopts the Tabular Prior Data Fitted Network (TabPFN) model which was originally designed to solve small tabular classification. TabPFN is used to predict the still or live birth during pregnancy with 97.91% accuracy. To address this life-saving problem with more accurate results and in-depth analysis of ML models, this research work makes use of 13 renowned ML models for performance comparison with the proposed model. The proposed model is evaluated using precision, recall, F-score, Mathews Correlation Coefficient (MCC), and the area under the curve evaluation parameters and the results are 97.87%, 98.26%, 98.05%, 96.42%, and 98.88%, respectively. The results of the proposed model are further evaluated using k-fold cross-validation and its performance comparison with other state-of-the-art studies indicating the superior performance of TabPFN model.

Just Published
Relevant
AI driven interpretable deep learning based fetal health classification

In this study, a deep learning model is proposed for the classification of fetal health into 3 categories: Normal, suspect, and pathological. The primary objective is to utilize the power of deep learning to improve the efficiency and effectiveness of diagnostic processes. A deep neural network (DNN) model is proposed for fetal health analysis using data obtained from Cardiotocography (CTG). A dataset containing 21 attributes is used to carry out this work. The model incorporates multiple hidden layers, augmented with batch normalization and dropout layers for improved generalization. This study assesses the model's interpretation ability in fetal health classification using explainable deep learning. This enhances transparency in decision-making of the classifier model by leveraging feature importance and feature saliency analysis, fostering trust and facilitating the clinical adoption of fetal health assessments. Our proposed model demonstrates a remarkable performance with 0.99 accuracy, 0.93 sensitivity, 0.93 specificity, 0.96 AUC, 0.93 precision, and 0.93 F1 scores in classifying fetal health. We also performed comparative analysis with six other models including Logistic Regression, KNN, SVM, Naive Bayes, Random Forest, and Gradient Boosting to assess and compare the effectiveness of our model and the accuracies of 0.89, 0.88, 0.90, 081, 0.93, and 0.93 were achieved respectively by these baseline models. The results revealed that our proposed model outperformed all the baseline models in terms of accuracy. This indicates the potential of deep learning in improving fetal health assessment and contributing to the field of obstetrics by providing a robust tool for early risk detection.

Open Access Just Published
Relevant
Relationship between asymmetry of transverse sinus and difference in intraocular pressure Based on MRV imaging examination

Transverse sinus asymmetry refers to the inconsistencies in the shape structure, size or blood flow of the intracranial transverse sinus. Intraocular pressure difference refers to the obvious difference in intraocular pressure between the two eyes. Transversal sinus asymmetry may be correlated with intraocular pressure difference, but the mechanism of correlation is still unclear. To investigate the relationship between transverse sinus asymmetry and IOP differences based on MRV examination, and to explore the possible mechanism. Patients with transverse sinus asymmetry were selected and examined using the MRV technique. At the same time, the patients' IOP was measured using standard methods of IOP measurement. Correlation analysis and statistical methods were used to evaluate the association between transverse sinus asymmetry and IOP differences. There was a statistically significant distinction observed between groups I and V (Z=6.78, P<0.01). Significant variations were also noted in the intraocular pressures across all groups, encompassing the average measurements of the right eye and left eye, along with the variance between the two (right eye: F=15.43, P<0.01; left eye: F=4.62, P=0.002; variance between eyes: F=41.79, P<0.01). The asymmetry of the transverse sinus exhibited a negative relationship with the intraocular pressure of the right eye (r=0.51, P<0.01) and the difference between the pressures of the two eyes (r=0.79, P<0.01); no significant association was found between the asymmetry and the left eye's intraocular pressure. In conclusion, a certain correlation exists between the intraocular pressures of the left and right eyes and the morphology of the transverse sinus. When the transverse sinus is thicker on one side, the corresponding drainage veins are thicker, resulting in lower intraocular pressure on that same side.

Open Access
Relevant
Bio Inspired Technological Performance in Color Doppler Ultrasonography and Echocardiography for Enhanced Diagnostic Precision in Fetal Congenital Heart Disease

The aim of this experiment is to investigate the bioinspired diagnostic performance of color Doppler ultrasound (CDUS) and two-dimensional (2D) echocardiography (ECG) for fetal congenital heart disease (FCHD). The research subjects were 33 expectant mothers with a diagnosis of FCHD at Xiangyang No. 1 People's Hospital between January 2017 and January 2021. The accuracy, sensitivity, and specificity of the two detection techniques were computed to ascertain and compare the diagnostic efficiency after CDUS and ECG examinations of all pregnant women. According to the findings, the prenatal CDUS detection rate was 92.61% higher than the 2D ECG detection rate (64.32%). The CDUS had an accuracy of 93.94%, sensitivity of 93.55%, and specificity of 100%, detecting 29 true positives, 0 false positives, 2 false negatives, and 2 true negatives. At 84.85% accuracy, 88.89% sensitivity, and 80% specificity, the 2D ECG identified 16 true positives, 3 false positives, 2 false positives, and 12 true negatives. There was a statistically significant (P < 0.05) difference between the accuracy, sensitivity, and specificity of 2D ECG and CDUS. In summary, CDUS was more effective than 2D ECG in diagnosing prenatal FCHD, and it also had a lower rate of missed and incorrect diagnoses.

Open Access
Relevant
HTS Library Plate Rejuvenation Using a DMSO-Rich Atmosphere

Dry DMSO can rapidly pull water vapor out of the air due to its hygroscopic nature. This is a well-documented problem within drug discovery, particularly within high-throughput screening (HTS). This hydration is caused by atmospheric moisture being absorbed each time a compound library is used. This effect becomes increasingly pronounced when a compound library is used routinely. The result of this hydration is a change to both the total volume of solution and the concentration of sample still in solution. This can result in a large amount of variability in the measured biological activity of a sample depending on the library usage. In this paper, we show the detrimental effects the hydration of sample libraries has on the reproducibility of biological data and present a novel way to remove it from HTS library plates. Our approach involves creating a DMSO-rich environment, created by placing anhydrous DMSO in compound storage pods purged with nitrogen, and incubating library plates in this environment for up to 3 days. Quantification via evaporative light scattering detection (ELSD) showed that removing water greatly increased the molarity of solutions, with a greater effect being seen for compounds with poor solubility. We also demonstrated how this approach can restore the inhibitory activity of stock solutions of compounds (pIC50) of samples containing ∼30% water from >30 µM to sub-micromolar after moisture removal. This method improves the reliability of tested compounds in HTS by potentially saving pharmaceutical companies hundreds of thousands of dollars in screening campaigns and increasing the quality of data.

Open Access
Relevant
Identification of m6A-related lncRNAs prognostic signature for predicting immunotherapy response in cervical cancer

BackgroundN6-methylandenosine-related long non-coding RNAs (m6A-related lncRNAs) play a crucial role in the cancer progression and immunotherapeutic efficacy. The potential function of m6A-related lncRNAs signature in cervical cancer has not been systematically clarified. MethodsRNA-seq and the clinical data of cervical cancer were extracted from The Cancer Genome Atlas. All of the patients were randomly classified into training and testing cohorts. The m6A-related lncRNAs prognostic model was constructed by LASSO regression using data in the training cohort.The predictive value of the signature was validated in the whole cohort and testing cohort. Cervical cancer patients were divided into low- and high-risk subgroups by the median value of risk scores. Kaplan-Meier analysis, principal-component analysis (PCA), functional enrichment annotation, and nomogram were used for further evaluation. We also examined the immune response and potential drug sensitivity targeting this model. ResultsSeventy-nine prognostic m6A-related lncRNAs were screened. The risk model comprising four m6A-related lncRNAs (AL139035.1, AC015922.2, AC073529.1, AC008124.1) was identified and verified as an independent prognostic predictor of cervical cancer. A nomogram based on age, tumor grade, clinical stage, TNM stage, and four m6A-related lncRNAs risk signatures was generated. It displayed good accuracy and reliability in predicting the overall survival of patients with CC. Based on our risk model, cervical cancer patients with potential immunotherapy benefits from the candidate drugs could be effectively screened. ConclusionThe four m6A-related lncRNAs signature may provide new targets and allow the prediction of immunotherapy response, which can assist developing individualized treatment for cervical cancer.

Relevant